I'm looking forward to the rest of the series. It's a nice departure from the usual (although still useful and definitely not mundane) stuff we all read on here.

I will have to correct you about the non-clustered indices though. They don't store the primary key in the leaf nodes. They actually store the clustered index field values in the leaf nodes. Often, or at least by default, the pkey and the clustered index are the same. If they are not the same then you may wonder what would happen if the clustered index is no longer unique... In that case MS SQL actual adds what is called a uniqifier to make the record, as far as the clustered index is concerned, unique. There's an article somewhere on this site which explains it well and in depth.

Apart from that, as I said, I'm looking forward to see how you develop your engine, etc. When I was younger I often wanted to write my own Dune 2000 clone. I never even got as far as you have in your first article! 😀